What happened to "CCD is better than CMOS"?!

[epines]

Sure from a technological point of view CMOS has surpassed CCD - but doesn't mean it's the best choice for everyone.

CCD

a. is cheaper
b. is more readily available
c. comes in a lot of varieties up to 80mpx
d. is tested and proven and reliable technology
e. still delivers great image quality and great resolution and DR
(for P1 only: f. has Sensor+which is a nice feature.)

A photographer who works mainly inside a studio has no real need for a great high ISO performance or an additional stop of DR but he may want to also buy a tech cam for product shots and another powerful flash head and use the highest resolution he can afford.

Right now CMOS is better from a technological point of view, but CCD sensors are also excellent and they are cheaper especially when it comes to 2nd hand market - and that's a big deal. Personally I couldn't afford a CMOS DB and I have no need for it.
So for me the CCD is "better".

Agreed. While the latest CMOS back costs in the $40K range, I got a brand-new H5D-50 system for a fraction of that. It's extremely well-made and blows away anything else I've used. True Focus is a godsend. Lenses are great and easily gotten in the second-hand market. No dark frame required on a tech camera. Yes, it gets noisier at higher ISOs. So what? Be a pro and light your shot, including the location. Frankly, using more lighting rather than relying on available light has helped bring my work to the next level and look different from the competition. And CCD noise at high ISOs, if you go that route, isn't bad looking, and doesn't even show up unless you're printing quite large.

The former top cameras in the world don't suddenly become irrelevant because something newer comes out. That's the technology trap that the manufacturers want you to fall into. CMOS sensors are great tools, but so are CCDs.

[voidshatter]

Agreed. While the latest CMOS back costs in the $40K range, I got a brand-new H5D-50 system for a fraction of that. It's extremely well-made and blows away anything else I've used. True Focus is a godsend. Lenses are great and easily gotten in the second-hand market. No dark frame required on a tech camera. Yes, it gets noisier at higher ISOs. So what? Be a pro and light your shot, including the location. Frankly, using more lighting rather than relying on available light has helped bring my work to the next level and look different from the competition. And CCD noise at high ISOs, if you go that route, isn't bad looking, and doesn't even show up unless you're printing quite large.

The former top cameras in the world don't suddenly become irrelevant because something newer comes out. That's the technology trap that the manufacturers want you to fall into. CMOS sensors are great tools, but so are CCDs.

Even DPReview has criticized the DR of the Canon 5DSR because sometimes it's not possible to have DR within control even if you own the most expensive Broncolor.

[landscapephoto]

Agreed. While the latest CMOS back costs in the $40K range, I got a brand-new H5D-50 system for a fraction of that. It's extremely well-made and blows away anything else I've used. True Focus is a godsend. Lenses are great and easily gotten in the second-hand market. No dark frame required on a tech camera. Yes, it gets noisier at higher ISOs. So what? Be a pro and light your shot, including the location. Frankly, using more lighting rather than relying on available light has helped bring my work to the next level and look different from the competition. And CCD noise at high ISOs, if you go that route, isn't bad looking, and doesn't even show up unless you're printing quite large.

The former top cameras in the world don't suddenly become irrelevant because something newer comes out. That's the technology trap that the manufacturers want you to fall into. CMOS sensors are great tools, but so are CCDs.

I could not agree more. H4D-50 here. 

[epines]

Even DPReview has criticized the DR of the Canon 5DSR because sometimes it's not possible to have DR within control even if you own the most expensive Broncolor.

Absolutely. Not surprising. And that's a CMOS sensor. In a case like this, if you really want to shoot with the sunset behind the talent and you want nice detail everywhere, the solution is to stay on the tripod, shoot plates at varying exposures, and composite in post.

[voidshatter]

Absolutely. Not surprising. And that's a CMOS sensor. In a case like this, if you really want to shoot with the sunset behind the talent and you want nice detail everywhere, the solution is to stay on the tripod, shoot plates at varying exposures, and composite in post.

That's a CMOS sensor from Canon, which is known to have limited DR as the CCD sensors. If you shoot with a CMOS sensor from Sony (e.g. IQ3 100MP, IQ3 50MP, Nikon D810 etc) then you would have a huge room for shadow recovery in post-processing.

[Paul2660]

That's a CMOS sensor from Canon, which is known to have limited DR as the CCD sensors. If you shoot with a CMOS sensor from Sony (e.g. IQ3 100MP, IQ3 50MP, Nikon D810 etc) then you would have a huge room for shadow recovery in post-processing.

Thanks you beat me to it. Canon CMOS to my eyes is still quite lacking in overall DR. I realize there are ways around this such as bracketing or lighting the subject et all.

CMOS works for some others not. For me it works. Just wish the cost of entry from Phase was a bit more dollar friendly. But as for now they are the only game in town and as such can charge what they like.

Paul C

[Jeffery Salter]

That's a CMOS sensor from Canon, which is known to have limited DR as the CCD sensors. If you shoot with a CMOS sensor from Sony (e.g. IQ3 100MP, IQ3 50MP, Nikon D810 etc) then you would have a huge room for shadow recovery in post-processing.

Or.  And there always is one. 

Simply create an image that shadows are a creative element and part of the composition. Yes its prudent that a craftsman understands a camera's inherent limitations (such as Dynamic range)  but probably not to the extent that the artist inside spends all the time pixel baiting rather than picture making.

This is not directed at Yunli Song or anyone in particular. Just a random off-topic thought.  Its a fruitless idea of arguing for limitations and not nurturing the imagination.

[Theodoros]

Canon sensors aren't worst than Sony for DR... nor Sony (FF - don't know about the MF ones, haven't tried one) sensors are anywhere near CCD MF sensors... As DR extension can only be considered (by a photo-grapher) what you are left to print with after processing... It all then depends if one wants his whites as being ...whites, or if he prefers a "brown sugar" white look...

[eronald]

Burnouts are also a creative element, but they work better on film than digital.

Edmund

Or.  And there always is one. 

Simply create an image that shadows are a creative element and part of the composition. Yes its prudent that a craftsman understands a camera's inherent limitations (such as Dynamic range)  but probably not to the extent that the artist inside spends all the time pixel baiting rather than picture making.

This is not directed at Yunli Song or anyone in particular. Just a random off-topic thought.  Its a fruitless idea of arguing for limitations and not nurturing the imagination.

[Theodoros]

Burnouts are also a creative element, but they work better on film than digital.

Edmund

Burnouts depend on exposure not on media... Intentional burn outs look the same on film or CDD MF sensor... That's why MF CDDs are exposed for the highlights as film was... (by the knowledgeable).

[ErikKaffehr]

Hi,

In film days we exposed transparencies for highlight and negative films for shadows. But, deciding what highlight needed to be included in the picture was always a compromise.

Transparency film had 18% grey (*) something like 2.5 stops under saturation, but saturation was sort of smooth. To my best knowledge, camera sensors put 18% grey something like 3 stops under saturation, but with sensors saturation clips. In special, colours are distorted when the channels don't clip in sync.

The great thing with sensors is that we can underexpose 18% (*) and still have very good fidelity. If we underexposed Velvia two stops shadows would have gone black.

Shooting negative film was a different thing. Negative film was noisy in the shadows but had a very long slope in the highlight and a very high dynamic range. The shoulder part did not go into clipping but in slow saturation. So negative film was exposed for the shadows.

The other part of the game is that folks like Adobe developed tools handle the characteristics of digital sensor. So we got highlight recovery, shadows expansion, tone mapping and so on.

In film times we had quite a lot of control with black and white, using development times, different developers and graded papers. With colour less so, I guess.

With great sensors assisted by advanced raw developers in the front and Photoshop in the back end we have a much simpler life. But neither turns us into an Ansel Adams.

Best regards
Erik

Burnouts depend on exposure not on media... Intentional burn outs look the same on film or CDD MF sensor... That's why MF CDDs are exposed for the highlights as film was... (by the knowledgeable).

[Phil Indeblanc]

Without much testing but with lots of observation, the CCD colors are rich out of camera. There is something maybe insignificant to many, but I notice a different "smooth" yet detailed rendering of texture.....And perhaps it is easily created in post. :-)

I sold my other back and kept a H25. I can only see it vs the 5Dm2. So from the start there is something rendered a bit more pleasing to my eye. This could be a number of things like the lens difference, the AA filter difference as well.

So what I gather is that the more I can isolate the variables in my gear the more control it gives me. lenses I want with the back I want.
I would not purchase a back dedicated to one body in digital back setups.

When the A7R2 came out, I was just about to buy it. But work using a DSLR was slow with big ticket sales and thought the gear I have now/5Dm2 is doing great, so no need.
What I was hoping for is having a New field cam, AND one that would trump my H25 with one shot with the major savings from buying a MF DB.

Now that the dust has settled, and that CMOS 50mpixel backs are around 7-9K.....
I still wonder if the combo of Rod or Schneider HM/Dig lenses I use on a Sinar with a Hass 50c DB, will it be better than a A7R2 with a 180, or 100 Macro or my Leica lenses? Splitting some hairs...where would they fall? This I can only know if I test them side by side in the subject I shoot.

[Christoph B.]

In film days we exposed transparencies for highlight and negative films for shadows. But, deciding what highlight needed to be included in the picture was always a compromise.

Speaking of film and CCD vs CMOS and the dynamic range..is the difference really significant?: (DxO may be a bit biased but that's a good estimate of the capabilities of those cameras/backs)


PhaseOne IQ180
about 13.5 DR

Sony A7R II
about 13.9 DR

Nikon D810
about 14.8 DR


I think we're talking about numbers that simply don't mean a lot any more. Really high quality negative film has (or rather had) a DR of 12-13 max, slide film only about 7 - and we're talking about a difference of 0.4 or 1.3 (still well beyond a DR of 12) and call it 'bad'?

When is good enough good enough?
If it has real life consequences that's one thing, but I seriously doubt anyone of us ever thought "if only I had that liiittle bit of additional dynamic range that would look so much better".  In all honesty - who can truthfully claim to have missed or ruined a shot because of a lack of 0.4 or 1.3 DR?
I see so many examples of people going DR-mad and making everything into surreal HDRs that look horrible while really breathtaking landscapes show a much narrower overall dynamic range in the finished photo.

So why are there so many people who find it necessary to belittle technology for an 'advantage' they can't and don't even really use?

[ErikKaffehr]

Hi,

I would agree as I feel that we have ample DR, mostly. I have seldom felt me limited by DR.

I know a guy who organises workshops and he shoots both Nikon 810 and Canon 5D/5DsR. His take on the issue that he needs to bracket a bit more with the Canon and sometimes needs HDR, while with the Nikon he can just expose to the right. But, most of the time he shoots Canon.

I would add some remarks regarding the DxO data:

• The DR given is engineering DR, which allows for very noisy shadows.
• DxO normalises the DR-data for a given size print. If you would pixel peep or print larger the DR-s would be different

Bill Claff has calculated another DR which is based on a reasonable signal/noise ratio that he calls photographic DR:

Camera	Photographic DR (BClaff)	Pixel peeping DR (DxO screen mode)
Phase One IQ 180	10.8 (derived from DxO)	11.89
Phase One IQ260	9.99
Nikon D810	11.51	13.67
Sony a7rII	11.38	12.69

Some authors state that lens flare limit scene luminance range, when projected on sensor, to about 11 steps. I don't think it is correct, but I would say it is a good assumption in most cases. A typical case where luminance range can be excessive is when a dark church is illuminated by small windows and you want to keep detail in those windows.


Here is an example from my P45+:


A HDR shot with my P45+ (Using Lumariver HDR):


A single exposure with my Sony Alpha 99:


The whole image


Best regards
Erik

Speaking of film and CCD vs CMOS and the dynamic range..is the difference really significant?: (DxO may be a bit biased but that's a good estimate of the capabilities of those cameras/backs)


PhaseOne IQ180
about 13.5 DR

Sony A7R II
about 13.9 DR

Nikon D810
about 14.8 DR


I think we're talking about numbers that simply don't mean a lot any more. Really high quality negative film has (or rather had) a DR of 12-13 max, slide film only about 7 - and we're talking about a difference of 0.4 or 1.3 (still well beyond a DR of 12) and call it 'bad'?

When is good enough good enough?
If it has real life consequences that's one thing, but I seriously doubt anyone of us ever thought "if only I had that liiittle bit of additional dynamic range that would look so much better".  In all honesty - who can truthfully claim to have missed or ruined a shot because of a lack of 0.4 or 1.3 DR?
I see so many examples of people going DR-mad and making everything into surreal HDRs that look horrible while really breathtaking landscapes show a much narrower overall dynamic range in the finished photo.

So why are there so many people who find it necessary to belittle technology for an 'advantage' they can't and don't even really use?

[AreBee]

Christoph_B,

When is good enough good enough?

When it is no longer not good enough.

[Christoph B.]

Hi,

I would agree as I feel that we have ample DR, mostly. I have seldom felt me limited by DR.

I know a guy who organises workshops and he shoots both Nikon 810 and Canon 5D/5DsR. His take on the issue that he needs to bracket a bit more with the Canon and sometimes needs HDR, while with the Nikon he can just expose to the right. But, most of the time he shoots Canon.

Bill Claff has calculated another DR which is based on a reasonable signal/noise ratio that he calls photographic DR:

Sure the 5Ds/r has a rather 'low' DR of about 12.4, that could be critical in some moments - but I highly doubt Bill Claffs "calculations" of photographic DR especially when he says the IQ180 has a better DR-performance than the 260. (edit: or was that figure of the 180 just directly taken from DxO? It's not really clear)
 That's just nonsense, they should be at least equal and maybe with a slight advantage for the 260 because of its pixel size (not much but a little).

Besides , his definition and calculation of "photographic DR" is completely arbitrary and has no comparative basis with film photography so I don't think that would be a good argument against my last comment.

[eronald]

Here is the only test I have ever seen where you can see the effects of S/N in color discrimination without numerical pixel peeping. It's CFV50C against Sony A7R2, and somehow the Hassy wins very convincingly.

You need to zoom the images of cards with words printed on them and try to read the sentence ...

http://www.revoirfoto.com/pr/index.php?pg=128&c=4&lg=

Edmund

[ErikKaffehr]

Hi,

Bill Claff does his own measurement based on the masked pixels of the sensor. DxO measures DR a different way, probably based on an ISO-standard, described here: http://dougkerr.net/Pumpkin/articles/ISO_Dynamic_range.pdf

Unfortunately, DxO only published measurements for very few MFDs and none of present generations.

Best regards
Erik

Sure the 5Ds/r has a rather 'low' DR of about 12.4, that could be critical in some moments - but I highly doubt Bill Claffs "calculations" of photographic DR especially when he says the IQ180 has a better DR-performance than the 260. (edit: or was that figure of the 180 just directly taken from DxO? It's not really clear)
 That's just nonsense, they should be at least equal and maybe with a slight advantage for the 260 because of its pixel size (not much but a little).

Besides , his definition and calculation of "photographic DR" is completely arbitrary and has no comparative basis with film photography so I don't think that would be a good argument against my last comment.

[ErikKaffehr]

Hi,

The CFV50C is something like 70% larger in surface. Both sensors are same generation, so largest sensor wins...

I would like to see those raw files, though…

Personally, I seldom shoot above 100 ISO, but I feel that going up in ISO sharpening and noise reduction needs to balanced.

Best regards
Erik

Here is the only test I have ever seen where you can see the effects of S/N in color discrimination without numerical pixel peeping. It's CFV50C against Sony A7R2, and somehow the Hassy wins very convincingly.

You need to zoom the images of cards with words printed on them and try to read the sentence ...

http://www.revoirfoto.com/pr/index.php?pg=128&c=4&lg=

Edmund

[bjanes]

Besides , his definition and calculation of "photographic DR" is completely arbitrary and has no comparative basis with film photography so I don't think that would be a good argument against my last comment.

I agree with Eric and you that currently available DR is usually adequate for most shots, but in some cases the highest available DR is desirable. The definition of photographic DR must be arbitrary since the noise floor for acceptable DR varies with the user, but Bill Claff's definition is reasonable and is supported by such forum gurus as Jim Kasson and Jack Hogan.

In any case, large sensors gain in photographic DR calculations over engineering DR, since the higher noise floor is moved up towards the shot noise region from the read noise region. Until recently MFDBs with their high read noise CCDs suffered. The new large format CMOS sensors should have very high photographic DR. Don't look a gift horse in the mouth. 

Bill